Using biodiversity features to promote an Ecosystem-Based Conservation framework in terrestrial ecosystems

PLOS One | https://doi.org/10.1371/journal.pone.0336705 November 21, 2025 7 / 18 At the second-level screening, we reviewed the title, abstract, methods, and results sections, while the introduc- tion and discussion sections were not considered. Each study was initially assigned to a terrestrial ecosystem based on the search string under which it was retrieved. If the reported study area did not match the ecosystem implied by the search string, the study was reassigned. Final assignments were verified using the study’s area description (e.g., habitat, geographic location, coordinates, or maps) provided in the methods or results sections. Full eligibility criteria were then applied; studies were included if they reported field-based primary data, presented in absolute numbers on biodiversity distribution or species richness within Tanzania’s terrestrial ecosystems. Stud- ies were excluded if they were conducted outside Tanzania, focused on aquatic systems or crops, relied solely on molecular/genetic approaches or modelling without field data, or presented non-primary literature such as reviews (Fig 2). From the final pool of eligible studies, biodiversity data were extracted in a standardised manner. When article reported aggregated counts of species (e.g., insects, snails, birds, reptiles, amphibians, trees, bryophytes), these were summed within the appropriate biodiversity feature category: vertebrates, invertebrates, or plants. When species-level inventories

were provided, species were counted individually, and the total count was assigned to their respective taxonomic group. Therefore, the number of biodiversity features reported in this study represents the aggregate totals of species within the three key taxonomic groups (plants, vertebrates, and invertebrates) as presented in each reviewed article, rather than unique species inventories. The number of articles contributing data for each taxon is summarised in Supplementary Materials (S1 Table). For each study, we documented the total counts of biodiversity features reported within each taxon and ecosys- tem. These counts, whether presented as individual species or aggregated groups (e.g., number of bird species), were summed across all included studies to obtain totals per taxonomic group and ecosystem. The specific counts extracted from each article are provided in the supplementary materials (S2 Data). The combined sum of reported biodiversity fea- tures from all three taxonomic groups within each ecosystem was expressed as a number per 100 km2 [53]. In addition, the summed totals for each taxonomic group within each ecosystem were converted into percentages to enable compari- sons across taxa and ecosystems. While biodiversity is inherently complex and difficult to quantify precisely, our approach provides a practical framework for ecosystem-level assessment. We adopted the CBD (1992) definition of biodiversity, which encom- passes diversity within species, between species and of ecosystems, allowing us to capture representative infor-

mation necessary for developing and improving conservation strategies at ecosystem-scale. The articles that were reviewed were used to estimate the presence or absence of key biodiversity features within each ecosystem. Although the exact locations of individual species were not specified in the studies, the species were assumed to occur across various habitat areas within the respective ecosystem. Whilst this approach does not provide exact estimates of species richness, it serves as a pragmatic proxy for assessing biodiversity distribution at ecosystem-level. Results Protected and unprotected areas in different ecosystems The scale of Tanzania’s terrestrial ecosystems are summarised along with the proportions of protected and unprotected areas within them (Table 1; Figs 1 and 3). Miombo woodland was the largest, covering 484,942 km² (55%), while moor- land being the smallest, covering 1,426 km² (0.2%) (Fig 1). The total amount of protected land varied, with protected areas (PPAs and OPAs) covering 41% (358,528 km²) of the land, with the remaining 59% (523,361 km²) being unprotected (out- side formal designations). In grassland ecosystem, 89.2% of the land area was protected, whilst for mangrove ecosystem only 8% were under protection. For montane forest ecosystem less than a third of the total area was protected. Over half of the area in most ecosystems lacked protection with coastal, and montane forests particularly vulnerable with 77% and